Tag Archives: Valve

Posted on by

Two Lessons From Team Fortress 2

I’ve gotten sucked back in to Team Fortress 2 (TF2) lately and taking notes of the changes that have happened since I last played. In the course of poking around the Mann Co Shop I’ve been reminded that they have some pretty smart cookies over there at Valve and I for one welcome our new cookie overlords. Specifically, I’ve noticed two things in the store that capitalize on concepts I’ve written about here before.

For those of you who don’t remember, Team Fortress 2 is a class-based shooter that has gone free-to-play and supports itself through purchases in the Mann Co Store. You can buy different weapons and cosmetic items there, for prices ranging from just a few cents to …well, I’ll get to that in a second.

The first thing I noticed is that the store is now making an interesting use out of something called “the endowment effect.” I’ve written about it before, and the quick version is that once we feel we own something, we value it more. The classic experimental example was when researcher Richard Thaler gave subjects a coffee cup, then shortly afterwords asked how much they would be willing to sell it for. Relative to those who were not given a cup but instead asked how much they would pay for it to own it, those who already owned it placed a higher dollar value on the thing.

Now it's YOUR Backburner. Is it worth two bits to you NOW?

TF2 gets you to endow certain items from its store by letting you try them out, in game, for free. After seven days, the item goes away, but you’re given a chance to buy it. According to the endowment effect, people might value their new Backburner (level 10 flame thrower) more than they did before, and be willing to pay more for it. But Valve is nice enough and/or smart enough to know that they could probably really drive the bargain home by giving you a 25% discount on the newly tested item. So double whammy. You’re likely to value the item more, plus you don’t want to lose your chance at a discount.

Unfortunately Valve seems to only be doing this test drive system for items that are already pretty cheap –in the 50-cent range. I think they could get more use out of it if they let you try it with a few more expensive items. Maybe even put some kind of one-per-week limit on it to prevent doing too crazy, or reduce the test drive period for newer items. ((Or you know what? I’ve got pallates full of these ManCo Supply Crates that I’m never going to use because I’m not gonna pay $2 to play a TF2 lottery. I’d trade them for a test run of a new item.)) They could even capitalize on envious reactions from other players who see you using the new item.

And on the topic of “more expensive items,” we have our next psychology lesson from TF2. Here, look at this:

I know, right? Right?

Yeah, you’re reading that right. You can now buy a virtual diamond ring in Team Fortress 2 that you can rename and then gift to another player. For $100. ONE. HUNDRED. REAL. DOLLARS. This boggled my mind when I first saw it, but then I realized that they besides cashing in on a few big spenders, the developers may be aiming to capitalize on an age-old sales trick: the contrast effect. Again, this was one of the first topics I wrote about for this blog, but the quickie version is that our perceptions of price (or more to the point, value) can be changed if we see a super high price ((Or a very cheap price, for that matter.)) off the bat. Retail sales people use this trick all the time by showing you a more expensive suit first, which makes the cheaper items two racks over seem a lot more affordable, let alone accessories like socks or belts. Ever poked your nose into an upscale clothing store and seen some absurd, $2,000 handbag on display up front? Who would pay that? Well, that’s not the point. The point is that it’s there to make the $200 handbags on the table next to it seem a heck of a lot more affordable –more so than if the super expensive item weren’t there.

Some people are witty AND rich! Too bad they're apparently already engaged...

This is what I think Valve is doing with the diamond ring item. Sure, letting you rename the item, give it to someone, and then broadcast that transaction to the entire Team Fortress 2 community –the whole community that’s online at the time, not just people on your server– will lead some people with more money than sense to have some fun like in the screenshot above. ((That’s a real screenshot I captured myself during a match, by the way.)) But I think the real benefit is that next to a $100 ring, that $13 hat for your Demoman looks a lot more affordable.

Posted on by

The Psychological Weight of History

Despite a huge backlog of games trying to get my attention, I found myself playing a lot of Team Fortress 2 (TF2) lately. This is in part because of the loot system, which drops random items –mainly hats or weapons– for you to use in customizing your avatar. ((Hey, by the way, I have an article on the psychology of loot drops in an upcoming issue of GamePro. COINCIDENCE? You decide.)) This system has been in TF2 for a while, and it used to be that the only way of getting the gear you wanted was by getting it from a drop or by crafting it from raw materials (which also essentially came from drops). Many players rejoiced and were very proud of their silly hats and weapons.

Then, in late September 2010 Valve introduced the Mannco Store, which allowed you to buy –with real money– almost all of the items that you used to have to score from lucky drops. Players were suddenly faced with the prospect that the Kritzkrieg or Backburner that they had been running around with would now be indistinguishable from those bought from the store. Comments began to arise on message boards that this would lower the appeal of those original items, the implication being that the “pre-Mannconomy” versions should be worth more than the new copies readily available in the store.

Buy stuff AND get in fights? Double sold!

This made me think about a series of experiments performed by Paul Bloom and Bruce Hood, as related in the book, How Pleasure Works. ((Bloom, P. (2010) How Pleasure Works: The New Science of Why We Like What We Like. New York: W.W. Norton & Company.)) Bloom and Hood wanted to see if invoking an item’s history could make it more valuable. Capitalizing on a recent visit by the Queen of England to the town where the research was to take place, they brought in a bunch of six-year old children and showed them certain items (a spoon or a cup) that had supposedly belonged to the Queen. They then placed the items into a mock “duplicating machine” that would supposedly make copies of the items. The machine consisted of a pair of boxes with hidden doors in the back through which one could slip an item’s twin in order to trick the kids into thinking that the original item had been duplicated. You may think that this is selling the average six year old’s intelligence a bit short, but Bloom explains himself pretty well:

When we showed this machine to children, none thought it was a trick. This fits with other research that finds that children are perfectly credulous about unusual machines. There is no reason why they should be skeptical. They live in a world with giant flying canisters, metal-cutting laser beams, talking computers, and so on. And we already have a rudimentary two-dimensional duplicating machines –you can take a piece of paper with Michael Jordon’s autograph on it, put it in a photocopy machine, press the button, and end up with something indistinguishable from the original. What is so strange about a three-dimensional version of this? For the children we tested: nothing. ((ibid, page 109))

(Indeed, one could actually imagine doing this study using scanners and 3D printers without resorting to tricky. We live in the future, people. Let’s start using our technology to more thoroughly deceive our children in the name of science.)

The researchers then had children assign value to the original and duplicate items. Not surprisingly, they valued the original items much more highly because they had history and that history was seen as not transferable to the copies. To see if the same effect would happen when it was YOUR history that went with the object, Bloom and Hood conducted a follow-up study where they offered to duplicate kids’ security objects –for example, blankets or stuffed animals that some kids will never sleep without. Some kids refused to allow their special objects to be subjected to such shenanigans, but for those that did, the researchers offered to let them take home either the original or the copy. Almost all of them chose to keep the original.

Bloom argues that this is all because we perceive items as having essences based on their history or who they belonged to. I’ve written about the endowment effect, which causes us to value an object more once we own it. Similar thing. And I don’t know if Valve was thinking of blankies and teddy bears when they rolled out the Mannco store, but they did apparently realize that items with a history –that is, that were acquired from drops before they were available to buy– would be seen as more valuable and players would feel a sense of loss if it was suddenly considered –or perhaps more importantly seen as– equivalent to readily available duplicates.

It also has +10 psychological weight.

Their solution: put the word “Vintage” in front of the item. So “Force of Nature” is what you can buy from the Mannco Store or find via drops after the update. “Vintage Force of Nature” is the thing you’ve had all along. It’s different, even if it looks the same and acts the same and may have been owned by the Queen of England. I’m curious what would happen if Valve ran an experiment where they offered to buy back duplicate items and asked people what they’d sell them for. How much more would the “Vintage” versions of items be worth relative to the non-vintage? I’d guess a LOT more.

Posted on by

The Psychology of Apology (and Hugs)

I’m looking forward to next year’s Portal 2 by Valve, in no small part because of the co-op mode where you team up with another little robot buddy and make your way through test chambers. Mistakes are sure to be made, though, and you may end up flinging or dropping your comrade to his/her death. Or maybe crushing. Or burning. Burning is always a possibility. Burning is really unavoidable when you get right down to it.

Fortunately, Valve has included a variety of emotes for your little robots to share with each other, including one that says “I’m so sorry; let’s hug it out.”

Hugs

Cold, mechanical apologies. They work.

This got me thinking about some research I’ve read on the power of the apology and how it really is missing in a lot of games. In his book, The Upside of Irrationality, Dan Ariely describes a simple and interesting field experiment he and a colleague conducted to see what effect an apology would have on remedying a minor annoyance. They hired an actor to sit in a coffee shop and ask patrons to complete a short exercise in exchange for $5. I don’t even know what the exercise was –something about circling letters– but that’s not the point; the exercise was just there to take up 5 minutes of time. At the end of that time, the actor would come over and then pretend to overpay participants by “accidentally” mixing in a $5 bill with the singles he was supposed to give them. This was, of course, done so that the researchers could see how people reacted to being overpaid.

There were three experimental conditions. In the first, things happened just as I described above. In the second condition, the actor pretended to receive a cell phone call in the middle of explaining the task instructions, and yakked away like an annoying nitwit while the subject had to wait for him to finish the instructions. Not quite as bad as flinging the person into a thermal deterrence beam, but a little annoying. In the third condition, the actor also annoyed the participants by taking a phone call, but afterwords he immediately apologized.1

The results were that 45% of the people in the first, non-annoyed condition returned the extra money, thereby turning down a chance to hurt the experimenter.2 When the actor pretended to take a phone call in the middle of a conversation, only 14% of the people returned the extra money. Surprisingly, though, if he apologized after taking the call, the number of people who returned the extra cash was the same as those who had not been annoyed at all. As Ariely puts it, “1 Annoyance + 1 Apology = 0 Annoyance.”

Why? In a 1997 study3 Michael McCullough, Everett Worthington, and Kenneth Rachal found that a good apology forged forgiveness through the act of empathy –that is, understanding of emotions between the offended and the offender. Ironically, this is one reason why I think the little robot hugs in Portal 2 work so well: those little guys look like they’ll have a lot of personality and exhibit more emotion than avatars in most other games.

Of course, apologizing is possible wherever there’s voice or text chat, but it’s probably not used as often as it should be, and in a fast-paced multiplayer game with lots of things going on, it’s kind of hard sometimes to apologize or even be heard if you do. But with a slowly paced game like Portal 2 or Little Big Planet, there’s not only every chance to apologize for a goof up, but there’s a lot more riding on it in terms of how well the other person cooperates, communicates potential puzzle solutions, or even if he/she drops out of the game altogether. Fortunately, even simple –or even insincere– apologies are particularly potent.

And this is why I believe that hugging robots should be in every children’s Social Studies textbook. Make it happen, Congresspeople.